Processes hang in prune_dcache() when attempting to reclaim memory from dentry_cache
Issue
- Several processes appear to have a similar stack trace
bpbkar R running task 0 6921 1 7417 6837 (NOTLB)
ffff81017bbdbae8 0000000000000082 ffff810714d00780 ffff81063767e540
ffff8105c23197a0 0000000000000001 ffff8105c23197a0 ffff810625e2e080
00106019a96b16b8 0000000000003978 ffff8105c2319990 0000001780062ff8
Call Trace:
[<ffffffff8008e7fb>] __cond_resched+0x1c/0x44
[<ffffffff8003b2c3>] cond_resched_lock+0x33/0x49
[<ffffffff8002e7c4>] prune_dcache+0x2e/0x132
[<ffffffff800ebd54>] shrink_dcache_memory+0x19/0x33
[<ffffffff8003f46e>] shrink_slab+0xdc/0x153
[<ffffffff800cba74>] try_to_free_pages+0x1da/0x2d7
[<ffffffff8000f4b9>] __alloc_pages+0x1f8/0x308
[<ffffffff8002e151>] __alloc_skb+0x77/0x12d
[<ffffffff80026227>] tcp_sendmsg+0x564/0xb0b
[<ffffffff80037c11>] do_sock_write+0xc6/0x102
[<ffffffff80047604>] sock_aio_write+0x4f/0x5e
[<ffffffff800182c3>] do_sync_write+0xc7/0x104
[<ffffffff800a09d4>] autoremove_wake_function+0x0/0x2e
[<ffffffff800325f9>] sys_faccessat+0xe4/0x18d
[<ffffffff80016ab9>] vfs_write+0xe1/0x174
[<ffffffff80017373>] sys_write+0x45/0x6e
[<ffffffff8006149d>] sysenter_do_call+0x1e/0x76
- System hung in 100% system busy state
Environment
- Red Hat Enterprise Linux 5
- kernel version between 2.6.18-164.7.1.el5 and 2.6.18-194.31.1.el5
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.